INDEX

Explanations

terms indicating visibility and enabled state for UI components

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 polaire

-0.68

 betweenstory

-0.61

 oluşan

-0.60

Socks

-0.60

juta

-0.59

ManyToOne

-0.58

utsal

-0.58

camore

-0.57

 Polytechnic

-0.57

 maxX

-0.57

POSITIVE LOGITS

Enabled

2.30

enabled

2.02

 Enabled

1.93

 enabled

1.79

enable

1.75

 Enable

1.72

Enable

1.72

 ENABLE

1.69

 enable

1.63

ENABLED

1.63

Activations Density 0.075%

terms indicating visibility and enabled state for UI components

No Comments

No Known Activations